Comparison of the KF and Particle Filter Based Out-of-Sequence Measurement Filtering Algorithms

Current out-of-sequence measurement (OOSM) filtering algorithms belong to two distinct classes, Kalman filter (KF) or extended KF (EKF) based and particle filter (PF) based. This paper compares the performances of the multiple-lag KF and PF based OOSM filtering algorithms for a number of scenarios with linear dynamic and measurement models with additive Gaussian noises first. The KF with in-sequence measurements represents an optimal estimator. Therefore, for this case, we compare the performances of the OOSM filtering algorithms relative to the KF with in-sequence measurements. Numerical results show that the KF OOSM algorithm used in this study is optimal. Next, we evaluate the multiple-lag KF/EKF and PF based OOSM filtering algorithms using realistic Ground Moving Target Indicator (GMTI) sensor measurements. We use estimation accuracy and statistical consistency for comparison. Keyword: Out-of-sequence measurement (OOSM), Particle Filter (PF), Kalman Filter (KF), Extended Kalman Filter (EKF), KF/EKF based OOSM Algorithm, PF based OOSM Algorithm, Ground Moving Target Indicator (GMTI) Sensor.